Concern about the possible association of postmenopausal hormone use with risk of breast cancer has been increased by several recent studies. Little data are available on the long term effects of progestin used in conjunction with estrogen which has become so common in recent years. Moreover, because of the benefits of estrogen for other diseases, it is important to precisely quantify any risk related to breast cancer. We propose to extend our ongoing multicenter, population-based case-control study of breast cancer to evaluate the relationships between postmenopausal hormone use and alcohol consumption. We will specifically assess use of estrogens with progestins, duration and currency of hormone use, and the interrelationships with alcohol intake. We will interview 5,000 women with newly diagnosed breast cancer identified from the tumor registries of Wisconsin, Massachusetts, and New Hampshire, and 5,000 population-based controls drawn from driver's license and medicare lists. The process will be similar to our current study which achieved a response rate of about 90% for cases and controls and enrolled more than the projected number of subjects. By extending our current study now, we can take advantage of an efficient process developed over the past several years and a well trained team. The large sample is required because of the need for precise estimates for what likely will be modest associations and because progestin plus estrogen use of sufficient duration for study is still relatively uncommon. The University of Wisconsin Clinical Cancer Center and the Harvard School of Public Health will continue to coordinate the study and have submitted separate applications with identical research plans.